AMSTERDAM (ANP) - Nederland is na Duitsland de grootste Europese exporteur van apparatuur die de ontwikkeling en toepassing van kunstmatige intelligentie (AI) mogelijk maakt. Dat meldt kredietverzekeraar Atradius in een nieuw rapport. Ons land is goed voor 7 procent van de wereldwijde export van AI-gerelateerde goederen. De exportwaarde van ruim 80 miljard euro levert wereldwijd een elfde plaats op.

Nederland moet het vooral hebben van chipmachines en zijn logistieke functie. Dankzij ASML is Nederland wereldwijd de grootste exporteur van machines voor de productie van halfgeleiders. Daarnaast heeft Nederland een belangrijke logistieke rol bij de wederuitvoer van onder meer routers en netwerkswitches, schrijft Atradius.

De wereldwijde markt is "sterk geconcentreerd", merkt de kredietverzekeraar op. Zo zijn Aziatische economieën goed voor 65 procent van de AI-gerelateerde export. De vijftien grootste exportlanden vertegenwoordigen daarnaast tot 90 procent van de wereldwijde uitvoer.

Europa blijft daarmee voor veel cruciale onderdelen afhankelijk van een klein aantal landen in Azië en van de Verenigde Staten. "Juist nu AI snel wordt opgeschaald, maakt deze concentratie de keten gevoeliger voor geopolitieke spanningen en handelsverstoringen", stelt econoom Theo Smid van Atradius. "Voor Europa is het daarom cruciaal om de afhankelijkheid waar mogelijk te verkleinen en de eigen positie in cruciale onderdelen van de AI-keten te versterken." Dat kan bijvoorbeeld door meer diversificatie van leveranciers en investeringen in Europese technologische capaciteit, denkt Smid.

Voor de ontwikkeling en toepassing van AI zijn tal van goederen nodig, waaronder grondstoffen, processors, geheugenchips, servers, netwerkapparatuur en machines voor de productie van chips. De wereldwijde export van deze AI-gerelateerde goederen vertegenwoordigde in 2024 een waarde van ruim 2,9 biljoen euro. Dat is ongeveer drie keer zoveel als tien jaar geleden. Nederland staat met een exportwaarde van ruim 80 miljard euro wereldwijd op de elfde plaats.

JERUZALEM/ANKARA (ANP) - De aanvaarding door Syrië van Turkse troepen op een basis in het noordwesten van Syrië vormde een bedreiging. Dat argument voert Israël aan voor een aanval op de basis Abu Duhur in Syrië dinsdag.

"Israël heeft Syrië herhaaldelijk gewaarschuwd dat zo'n aanwezigheid een bedreiging zou zijn voor Israëls veiligheid", zei het kantoor van de Israëlische premier Benjamin Netanyahu in een verklaring. Volgens Israël heeft Syrië "ervoor gekozen deze waarschuwingen te negeren".

De Verenigde Staten hebben de Israëlische aanval op de basis in Syrië omschreven als een "onnodige escalatie". De Amerikaanse Syrië-gezant Tom Barrack schreef op X "diep bezorgd" te zijn over de aanval.

"Dit bevordert de regionale stabiliteit niet", aldus Barrack. Volgens de gezant heeft de Syrische regering de aanval niet verdiend, omdat ze zich niet vijandig heeft opgesteld en ook herhaaldelijk zou hebben aangegeven liever te de-escaleren. Barrack, die ook ambassadeur in Turkije is, wil dat de twee landen in gesprek met elkaar blijven.

Syrië en Turkije veroordeelden de Israëlische aanval ook. Volgens het Turkse ministerie van Buitenlandse Zaken zijn de Israëlische aanvallen een "schending van Syrië's territoriale integriteit". Het ministerie drong er bij de internationale gemeenschap op aan "sterker stelling te nemen om een eind te maken aan Israëls toenemende agressie tegen Syrië". Het ministerie zei niets over de aanwezigheid van Turkse troepen in Syrië.

Ankara heeft Israël er eerder van beschuldigd dat het Syrië destabiliseert en de nieuwe regering ondermijnt. Turkije is een van de belangrijkste bondgenoten geworden van de Syrische president Ahmed al-Sharaa en helpt bij de opleiding van het Syrische leger en bij de wederopbouw van overheidsinstellingen en infrastructuur.

Het kwam, slechts een enkele keer, weleens voor dat wij als wethouders iets wat die week niet helemáál lekker was gelopen zeg maar, dus ehhh… enigszins moeizaam formulerend, op…

Je krijgt een paar cijfers cadeau, maar het grid van 9x9 moet foutloos ingevuld worden.

Wie of wat staat er midden in het nieuws? Een actuele puzzel, die makkelijker is als je het nieuws een beetje volgt.

Een Precies Vier bestaat uit 16 woorden, begrippen of namen, die moeten worden verdeeld in precies vier groepen van vier. Er is telkens maar één oplossing mogelijk. Welke woorden vormen een connectie?

Een kruiswoordpuzzel, maar dan heel klein (en snel).

If high-speed AI inference is what you’re after, memory bandwidth is the bottleneck to beat. At a mind-numbing 21.6 petabytes per second (PB/s) of memory bandwidth, Cerebras' dinner-plate-sized AI accelerators were already 1,000x faster than Nvidia's or AMD’s best GPUs. The chip newcomer unveiled its next-gen Wafer Scale Engine (WSE) and Nexus rack systems on Tuesday. Cerebras aims to extend that lead by boosting throughput per watt tenfold over the previous generation. Putting the 'T' in Turbo Cerebras accomplishes this in a couple of ways. But, from what we can tell, the primary lever comes from squeezing its chips for every hertz they’ve got. The newly announced WSE-3T — the “T” here stands for “Turbo” — promises twice the compute, memory fabric, and I/O bandwidth of the now two-year-old WSE-3. Yet, if you look at the chart below, you’ll notice it accomplishes this using the same process tech, wafer area size, transistor count, core count, and SRAM capacity. That's because the WSE-3T isn't new silicon. Instead, Cerebras tells us it's just pushing its existing wafer scale engine harder. The main innovation this time around seems to be related to power delivery, which is apparently so efficient that they’re able to push twice the power through the chip, which “enables higher operating frequencies and faster token generation.” How much higher does it clock? By our estimate, Cerebras is now running the silicon at 2.8 GHz, up from 1.4 GHz last gen, which would be quite the accomplishment. In any case, each WSE-3T boasts 250 petaFLOPS of AI compute, 44 GB of SRAM (that’s not a typo, there really is that much SRAM on there), good for 43.2 PB/s of memory bandwidth, and 2.4 Tbps of off-die connectivity. On paper that sounds more impressive than it really is. AMD and Nvidia’s latest GPUs offer 4 to 5 petaFLOPS of dense FP16 compute or 35 to 50 petaFLOPS at FP4. Cerebras’ headline performance figure relies heavily on sparsity, which as a general rule doesn't benefit LLM inference. Assuming the same 10x sparsity we saw with the WSE-3, the WSE-3T’s dense FP16 performance should be closer to 25 petaFLOPS, which is still impressive, just not as impressive as the chipmaker would have you believe. We also suspect the WSE-3T’s peak memory bandwidth is purely theoretical. During LLM inference, the WSE-3 lacked the compute necessary to saturate its SRAM on its own, and we have no reason to believe the Turbo variant will be any different. However, this time around Cerebras isn’t trying to run the entire inference stack on its own accelerators. Instead, it has partnered with Amazon Web Services (AWS) and AMD to offload the compute-intensive prompt processing bits of the inference pipeline onto their respective Trainium XPUs and Instinct GPUs. At least for inference, Cerebras’ chips now function primarily as decode accelerators, similar to how Nvidia is using Groq — not to be confused with Elon Musk’s Grok family of models — LPUs in its LPX rack systems. The major benefit for Cerebras is its chips have a whack ton of SRAM on board. So, instead of needing 2,000 LPUs to run a trillion-parameter model, Cerebras can get away with using a few dozen, depending on the precision at which the weights are stored. Curiously, Cerebras opted to double performance this generation rather than boost SRAM capacity, which hasn’t increased meaningfully since the WSE-2 launched five years ago. In a disaggregated inference environment where prefill is handled by GPUs, we’d have expected to see Cerebras prioritize SRAM capacity over compute. However, given that these disaggregated compute architectures are a relatively new phenomenon, it’s possible Cerebras was already too far along in production to pivot. This likely explains the Turbo naming convention. If Cerebras plans to continue down this path, we expect the WSE-4, which is presumably still coming, to offer only modest performance gains at FP16 while roughly doubling SRAM capacity. Our sibling site The Next Platform has drawn up some predictions of what the WSE-4 might look like if you’re interested. Cerebras goes rackscale Cerebras' latest generation of wafer scale accelerators also sees the company get serious about rack-scale compute architectures. Much like Nvidia’s NVL72 and AMD’s Helios racks, Cerebras’ CS-4 makes the leap from a monolithic system to a modular architecture that breaks out compute, power delivery, and cabling for easier deployment, maintenance, and upgrades. The company’s chips are now housed in what it calls a “backpack” form factor, a sort of self-contained system with all the control electronics on board. Each CS-4 can be equipped with up to three of these backpacks, which, as their name suggests, plug into the back of the rack, while the front of the rack is dedicated to the power shelves used to feed the beast. Cabling presumably runs down the center of the system. We're guessing it's probably not an OCP-compliant design, but when your chip is the size of a charcuterie board, it never was going to be. With 3x the accelerators per system, power consumption has unsurprisingly increased. Cerebras hasn’t said how much power the racks will suck back, but it has said that its more efficient power delivery means it can push twice as many watts through the chip. The WSE-3 was already a hot chip at 15 kW at the wafer level and around 23 kW at the system level. This means we’re probably looking at around 46 kW for each CS-4 backpack and a total system power of between 120 kW and 140 kW. A few years ago, that’d have been a monstrous amount of power, even for a liquid-cooled machine. Today, the machine looks positively conservative next to the 240 to 250 kW rack systems coming from AMD and Nvidia later this year. Killing latency by ditching the switch Even with 132 GB of SRAM memory per rack, you’re still going to need a lot of racks to run any reasonably sized model, which means I/O is a major consideration. It just so happens that the new racks and silicon that power them have gotten some beefy upgrades in this respect. Each chip is equipped with 2.4 Tbps of chip-to-chip bandwidth, up from 1.2 Tbps. But arguably the more important improvement is to latency, which thanks to some clever tricks has been cut from five microseconds down to just two. That’s because unlike GPUs, which rely heavily on bandwidth-intensive tensor and expert parallelism to multiply their compute and effective memory bandwidth, Cerebras' chips are already so fast they can get away with using pipeline parallelism. Pipeline parallelism is about as simple as multi-accelerator inference gets. Model weights are distributed across each accelerator and work is performed sequentially, one chip after another, hence the name pipeline parallelism. While interconnect bandwidth isn’t as much of an issue, pipeline parallelism does benefit from very low latency connections. Cerebras is able to lower its interconnect latency in a pretty simple way: Get rid of all the extra switches and just have the chips talk to one another. This is a pretty big departure from what a lot of AI chip designers have been doing with their own rack-scale designs. AWS for example ditched its chip-to-chip mesh in favor of a switched fabric in its Trainium3 accelerators, which we looked at in detail late last year. Cerebras tells us it's using a 2D torus, which you can think of as a grid where the ends wrap around to the other side. The topology, the company says, can support models up to 50 trillion parameters in size, though no such model currently exists to our knowledge. And for models that do exist, it appears the parts will be quite performant, achieving speeds of up to 4,400 tok/s per user in gpt-oss-120b on a single CS-4 system, compared to around 350 tok/s on the fastest GPU-based inference service today, according to the benchmarking gurus at Artificial Analysis. Cerebras' new mesh topology isn’t mandatory. If you wanted to connect the chips using a switched fabric, there’s nothing stopping you. Like basically every modern chip, it supports RDMA over converged Ethernet (RoCE), too. The compromise is your latencies may be a bit higher. The chip upstart expects the first CS-4-based systems to come online later this quarter. ®

OpenAI on Tuesday said its decision to suspend model training work, implemented after unreleased, unsupervised AI models hacked HuggingFace, remains in effect as the AI biz tries to implement stronger security measures. Some of those measures will increase compute overhead by 20 percent of the observed inference workload. An OpenAI spokesperson told The Register that those costs reflect internal research and won't be passed on directly to customers. The company has not revealed what portion of its total inference compute is subject to such monitoring now, or under its prior monitoring regime. "We have paused some frontier RL [reinforcement learning] training to ensure that we can meet the appropriate alignment, security and monitoring standards for the new level of capabilities in front of us," OpenAI CEO Sam Altman wrote in a social media post. "Model progress is now extremely rapid, and we always said we would take action if we felt that model capabilities were outstripping the pace of safety and alignment." Altman said he still expects new models, presumably the delayed Astra, will ship soon. The training pause affects further-out releases. OpenAI in its post reiterated its plans to focus on monitoring, model alignment, and security measures to prevent its models from running amok as they did last month. Following the HuggingFace incident, OpenAI "paused frontier model inference in research clusters for runs that could execute code or use tools that could access the internet." The biz said it allows some workloads to run, but paused others until they can be moved under a more stringent security regime that includes sandboxing, network isolation, and continuous security testing. "Our largest planned frontier RL (reinforcement learning) run remains on hold while we conduct smaller-scale training and evaluations to assess model behavior, validate our safeguards, and establish more evidence of alignment before proceeding," the company wrote. Reinforcement learning refers to the trial-and-error process by which AI agents "learn" about their environment by being rewarded for desired outcomes. OpenAI also said it is expanding its monitoring of the chain-of-thought process, the technique that sees "thinking" models break down tasks into discrete steps and produce intermediate text output for each step. The company's prior approach focused on high-risk workloads, specifically internal deployments of frontier models and frontier RL training runs. In contrast, OpenAI says, its new monitoring setup covers all RL training and evaluations involving tools for models at the capability level of GPT-5.6 Sol or higher. And with the determination that Astra possesses critical cyber capabilities, OpenAI added an additional monitoring requirement that covers all inference with Astra, not just RL training and testing. "These safeguards require meaningful compute," OpenAI said. "Our current estimates put monitoring overhead at roughly 20 percent of the inference compute being monitored, though the cost varies substantially across training and evaluation workloads." OpenAI expects to share more details about the implementation of its monitoring scheme in a future post. In research published last year, the company said that chain-of-thought monitoring is an effective way to detect model misbehavior, but cautioned that directly optimizing models to strictly follow instructions "does not eliminate all misbehavior and can cause a model to hide its intent." If you choose to believe the company's assurance that it will not pass on the cost of model thought policing to customers, it follows that OpenAI's losses will increase. It's difficult to imagine that would be a sustainable stance if OpenAI goes public. But given the company's reported $600+ billion in AI infrastructure commitments and its expectation to remain unprofitable until at least 2030, what's a bit more expense for the sake of uncertain security? ®

Thomas Hawk posted a photo:

Thomas Hawk posted a photo:

Thomas Hawk posted a photo:

Thomas Hawk posted a photo:

Thomas Hawk posted a photo:

Weer voor Rotterdam ☁️ - 19-08-2026 01:15 CEST

In één oogopslag:
• 19.8°C · Bewolkt ☁️ | Min 16.7°C / Max 21.5°C | Kans op neerslag 62%

Verwachting voor vandaag:
• Min 16.7°C, Max 21.5°C (Lichte regen) 🌧️, Neerslag 6.1 mm, Kans op neerslag 62%, 🧭 1005.2 hPa ↘️ -2.4 hPa/24h, Windsnelheid: 21.6 km/u (6.0 m/s), richting: → 253°

Uurlijkse voorspelling voor de komende 12 uur:

• 02:00: 19.7°C (Bewolkt) ☁️, Kans op neerslag 8%, 🧭 1007.6 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 5.8 km/u (1.6 m/s), richting: ↑ 175°
• 03:00: 19.5°C (Bewolkt) ☁️, Kans op neerslag 5%, 🧭 1007.3 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 7.6 km/u (2.1 m/s), richting: ↗ 204°
• 04:00: 19.4°C (Bewolkt) ☁️, Kans op neerslag 5%, 🧭 1006.4 hPa ↘️ -0.9 hPa/1h, Windsnelheid: 6.5 km/u (1.8 m/s), richting: ↗ 204°
• 05:00: 19.2°C (Bewolkt) ☁️, Kans op neerslag 14%, 🧭 1006.2 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 4.0 km/u (1.1 m/s), richting: ↑ 188°
• 06:00: 18.9°C (Lichte motregen) 🌦️, Neerslag 0.1 mm, Kans op neerslag 38%, 🧭 1005.5 hPa ↘️ -0.7 hPa/1h, Windsnelheid: 5.0 km/u (1.4 m/s), richting: ↗ 213°
• 07:00: 18.5°C (Lichte regen) 🌧️, Neerslag 1.9 mm, Kans op neerslag 70%, 🧭 1005.1 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 9.0 km/u (2.5 m/s), richting: ↗ 215°
• 08:00: 18.3°C (Zware motregen) 🌦️, Neerslag 1.2 mm, Kans op neerslag 94%, 🧭 1005.1 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 7.9 km/u (2.2 m/s), richting: ↑ 187°
• 09:00: 18.6°C (Lichte motregen) 🌦️, Neerslag 0.1 mm, Kans op neerslag 100%, 🧭 1005.0 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 13.0 km/u (3.6 m/s), richting: ↗ 212°
• 10:00: 19.3°C (Bewolkt) ☁️, Kans op neerslag 100%, 🧭 1005.4 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 19.8 km/u (5.5 m/s), richting: ↗ 241°
• 11:00: 17.5°C (Matige regen) 🌧️, Neerslag 2.9 mm, Kans op neerslag 98%, 🧭 1005.7 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 19.4 km/u (5.4 m/s), richting: → 285°
• 12:00: 17.8°C (Lichte motregen) 🌦️, Neerslag 0.4 mm, Kans op neerslag 87%, 🧭 1005.5 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 11.2 km/u (3.1 m/s), richting: → 272°
• 13:00: 17.9°C (Lichte motregen) 🌦️, Neerslag 0.3 mm, Kans op neerslag 72%, 🧭 1005.3 hPa ➡️ 0.0 hPa/1h, Windsnelheid: 9.7 km/u (2.7 m/s), richting: → 269°

Voorspelling voor de komende dagen:

• donderdag 20 augustus: Min 17.5°C, Max 19.8°C (Matige regen) 🌧️, Neerslag 10.9 mm, Kans op neerslag 50%, 🧭 1004.5 hPa ↘️ -0.7 hPa/24h, Windsnelheid: 19.8 km/u (5.5 m/s), richting: ↗ 226°
• vrijdag 21 augustus: Min 16.4°C, Max 20.6°C (Matige regen) 🌧️, Neerslag 12.9 mm, Kans op neerslag 84%, 🧭 1009.1 hPa ↗️ +4.6 hPa/24h, Windsnelheid: 27.4 km/u (7.6 m/s), richting: ↗ 219°
• zaterdag 22 augustus: Min 14.3°C, Max 19.0°C (Lichte onweersbui) ⛈️, Neerslag 15.4 mm, Kans op neerslag 63%, 🧭 1018.7 hPa ↗️ +9.6 hPa/24h, Windsnelheid: 15.8 km/u (4.4 m/s), richting: → 278°
• zondag 23 augustus: Min 12.7°C, Max 18.5°C (Zware motregen) 🌦️, Neerslag 2.8 mm, Kans op neerslag 29%, 🧭 1023.9 hPa ↗️ +5.2 hPa/24h, Windsnelheid: 12.8 km/u (3.6 m/s), richting: ↘ 329°
• maandag 24 augustus: Min 13.1°C, Max 18.8°C (Lichte motregen) 🌦️, Neerslag 0.6 mm, Kans op neerslag 7%, 🧭 1021.1 hPa ↘️ -2.8 hPa/24h, Windsnelheid: 9.7 km/u (2.7 m/s), richting: ↙ 25°
• dinsdag 25 augustus: Min 12.1°C, Max 19.3°C (Bewolkt) ☁️, Kans op neerslag 8%, 🧭 1013.4 hPa ↘️ -7.7 hPa/24h, Windsnelheid: 16.7 km/u (4.6 m/s), richting: ← 68°

Details:
• 🌡️ Huidige temperatuur (om 01:15): 19.8°C (Bewolkt)
• 🤚 Gevoelstemperatuur: 21.1°C (+1.3°C)
• 💨 Windsnelheid: 5.4 km/u (1.5 m/s), richting: ↘ 303°
• 🌬️ Windstoten: 8.6 km/h (2.4 m/s)
• 💧 Luchtvochtigheid: 82%
• 🧭 Luchtdruk: 1007.6 hPa ↘️ -1.2 hPa/3h
• 👁️ Zichtbaarheid: 13.3 km
• ☀️ UV-index: 0.0
• 🌅 Zonsopgang: 06:31 · 🌇 Zonsondergang: 21:00

Luchtkwaliteit:
• AQI: 37 🟢 (Goed)
• PM2.5: 10.4 μg/m³
• PM10: 14.2 μg/m³

Gegevens geleverd door Open-Meteo

#Rotterdam #Meteo #MeteoBot #MeteoNetherlands #OpenMeteo #FediMeteo